GECKOS-MUSE: Edge-on Disc Galaxy Survey

• GECKOS-MUSE is an ESO/VLT large programme that studies edge-on Milky Way-mass galaxies by integrating Galactic Archaeology with MUSE spectroscopy.
• The survey employs deep observations and advanced kinematic, chemical, and photometric analyses to resolve thin/thick disc structure, nuclear discs, and extraplanar ionized gas.
• Its results refine galaxy evolution models by linking stellar populations, disc dynamics, and multiphase outflows across both main-sequence and starburst systems.

GECKOS-MUSE is an ESO/VLT Large Programme built around the Multi Unit Spectroscopic Explorer (MUSE) to study edge-on disc galaxies with stellar masses close to that of the Milky Way, with the explicit aim of connecting Galactic Archaeology to integral-field spectroscopy of external systems. The survey name expands to “Generalising Edge-on galaxies and their Chemical bimodalities, Kinematics, and Outflows out to Solar environments,” and its core scientific scope spans thin/thick disc formation, chemical bimodality in $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$, bar- and bulge-related kinematic substructure, and extraplanar gas from diffuse ionised layers to galaxy-scale winds (Sande et al., 2023). In later GECKOS-MUSE papers, the programme is also described as a 317 h VLT/MUSE effort and as the first systematic, mass-controlled mapping of extraplanar diffuse ionized gas and winds in Milky Way-mass galaxies (Elliott et al., 30 Jan 2026).

1. Scientific rationale and edge-on perspective

The central ambition of GECKOS is to understand how the thin/thick disc dichotomy, the chemical bimodality seen in $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$, and the prevalence and impact of galactic outflows change across galaxies of Milky-Way mass but a wide range of assembly histories (Sande et al., 2023). This programme is motivated by the fact that high-resolution Milky Way surveys such as APOGEE, GALAH and Gaia have revealed an $\alpha$-rich, old thick disc distinct in age, scale-height and chemistry from the younger, $\alpha$-poor thin disc, while the origin of these two components—whether driven by mergers, clumpy early star formation, radial migration, internal heating, or some combination thereof—remains debated.

Edge-on discs are central to the survey design because they allow stellar light and ionised gas to separate naturally into midplane and off-plane components. In projection, the thick disc and any extraplanar ionised outflow stand apart from the bright thin disc, permitting measurements of gradients in velocity dispersion, mean stellar metallicity $[\mathrm{M}/\mathrm{H}]$, $[\alpha/\mathrm{Fe}]$, and emission-line ratios without the de-projection degeneracies that affect face-on or intermediate-inclination surveys (Sande et al., 2023). The same geometry also isolates boxy/peanut bulges, which are signatures of buckled bars and therefore a direct route to studying secular and bar-driven evolution of the inner disc.

A complementary motivation in the kinematic-substructure branch of GECKOS is to identify and quantify boxy/peanut bulges and their bar-driven kinematic signatures in two dimensions, search for nuclear discs via high-order kinematic moments, and test whether dispersion-dominated bulges are required by the data. In the first 12 galaxies observed, the survey reports that the kinematic maps of all sample galaxies can be explained via disc structure(s) alone, and that there is no need to invoke the existence of dispersion-dominated bulges (Fraser-McKelvie et al., 2024). A plausible implication is that GECKOS is designed not only to extend Milky Way diagnostics outward, but also to revise structural classifications inward, especially in dusty, edge-on systems where photometric decompositions alone are ambiguous.

2. Sample design, selection, and observational strategy

The survey overview defines a GECKOS sample of 35 nearby edge-on disc galaxies designed to trace assembly histories across a large range of star formation rates, bulge-to-total ratios, and boxy and non-boxy bulges (Sande et al., 2023). Later GECKOS-MUSE papers describe the parent programme as targeting 36 edge-on systems, including a first internal data release of 12 galaxies and a seven-galaxy JAM pilot subsample (Fraser-McKelvie et al., 2024); (Rutherford et al., 10 Sep 2025). The record therefore contains both 35- and 36-galaxy descriptions.

The main overview paper states the following quantitative cuts: distance $15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$, inclination $i>85^\circ$, stellar mass within $\pm 0.3$ dex of the Milky Way, specifically $9.7<\log(M_\star/M_\odot)<10.3$, star-formation rates spanning a 2 dex interval about the local main sequence, approximately $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$0 to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$1, bulge-to-total light ratios $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$2, an even mix of boxy versus round bulges, and avoidance of dense cluster environments where ram-pressure stripping dominates (Sande et al., 2023). The iDR1 kinematic paper reports target selection over $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$3, with edge-on or highly inclined systems at $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$4, stellar mass within $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$5 dex of $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$6, and a 2 dex span in mid-IR SFR (Fraser-McKelvie et al., 2024). The nine-galaxy extraplanar-gas subset is described as having distances $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$7–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$8, $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$9 within $\alpha$0 dex of the Milky Way, and SFRs from $\alpha$1 to $\alpha$2 (Elliott et al., 30 Jan 2026).

All GECKOS targets are observed with MUSE in Wide Field Mode on the VLT, with a $\alpha$3 field and $\alpha$4 spatial sampling (Sande et al., 2023). The wavelength coverage is reported as approximately $\alpha$5–$\alpha$6 or $\alpha$7–$\alpha$8, depending on the analysis paper, while the spectral resolution is given as nominally $\alpha$9, or more specifically from $\alpha$0 at $\alpha$1 to $\alpha$2 at $\alpha$3 in the JAM analysis (Rutherford et al., 10 Sep 2025). For stellar population fitting, one GECKOS description restricts the analysis to $\alpha$4–$\alpha$5 where the MILES templates are best calibrated (Sande et al., 2023).

The observing strategy is correspondingly deep. The overview paper states that to reach stellar $\alpha$6 at a surface brightness of $\alpha$7, integrations of up to $\alpha$8 minutes per galaxy are used, split into multiple observing blocks for cosmic-ray rejection and systematics control (Sande et al., 2023). The iDR1 kinematic paper describes 1 h blocks in an OSOOSO sequence, with four $\alpha$9–$[\mathrm{M}/\mathrm{H}]$0 on-source exposures and two $[\mathrm{M}/\mathrm{H}]$1–$[\mathrm{M}/\mathrm{H}]$2 sky frames, together with dithers and $[\mathrm{M}/\mathrm{H}]$3 rotations (Fraser-McKelvie et al., 2024). For the star-forming extraplanar-gas subset, a typical observing block comprises four $[\mathrm{M}/\mathrm{H}]$4–$[\mathrm{M}/\mathrm{H}]$5 min on-source exposures interleaved with offset sky frames, yielding $[\mathrm{M}/\mathrm{H}]$6 min on-source per pointing and $[\mathrm{M}/\mathrm{H}]$7–$[\mathrm{M}/\mathrm{H}]$8 pointings per galaxy (Elliott et al., 30 Jan 2026). A subset $[\mathrm{M}/\mathrm{H}]$9 dex above the main sequence receives additional off-plane pointings extending to $[\alpha/\mathrm{Fe}]$0 to capture any H$[\alpha/\mathrm{Fe}]$1 and [N II] emission from outflows (Sande et al., 2023).

3. Reduction pipelines, analysis framework, and data products

Raw MUSE exposures in GECKOS are reduced with the ESO MUSE pipeline, generally through pymusepipe, with the workflow including bias subtraction, flat-fielding, wavelength calibration, illumination correction, sky subtraction, flux calibration, astrometric alignment, and cube mosaicking (Elliott et al., 30 Jan 2026). In the extraplanar-gas analyses, individual cubes are aligned in WCS, stacked per observing block with MPDAF, and mosaicked into a final cube for each galaxy (Elliott et al., 30 Jan 2026). The NGC 4666 study additionally reports flux homogenisation across pointings by comparing disk spectra in overlap zones and applying a 3rd-order polynomial correction (Ciraulo et al., 22 Sep 2025).

GECKOS makes extensive use of GIST and nGIST for continuum fitting, kinematic extraction, and stellar-population analysis. In the survey overview, flux-calibrated cubes are further analysed with the GIST framework, producing Voronoi-binned stellar continuum maps at $[\alpha/\mathrm{Fe}]$2, stellar kinematics $[\alpha/\mathrm{Fe}]$3 extracted via pPXF fitting of MILES templates, mass-weighted and light-weighted age and metallicity maps, and emission-line flux maps corrected for underlying Balmer absorption (Sande et al., 2023). The kinematic-substructure paper presents nGIST as a modern pipeline developed to generate the specialised GECKOS data products, using Voronoi binning to $[\alpha/\mathrm{Fe}]$4, pPXF with the X-shooter stellar library DR3, and Monte Carlo error estimation with 100 perturb-and-refit realisations per bin (Fraser-McKelvie et al., 2024). The JAM pilot likewise uses nGIST with pPXF, X-Shooter templates, and Gaussian LOSVDs with $[\alpha/\mathrm{Fe}]$5 for the stellar-dynamical modelling stage (Rutherford et al., 10 Sep 2025).

The standard GECKOS emission-line set includes H$[\alpha/\mathrm{Fe}]$6, H$[\alpha/\mathrm{Fe}]$7, [O III] $[\alpha/\mathrm{Fe}]$8, [N II] $[\alpha/\mathrm{Fe}]$9, and [S II] $15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$0. In the extraplanar-gas work, these lines are fitted spaxel-by-spaxel with single Gaussians using threadcount, on both $15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$1 and 500 pc re-binned cubes, retaining only lines with flux $15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$2 and running ten Monte Carlo iterations for $15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$3 (Elliott et al., 30 Jan 2026). Internal extinction is corrected from the Balmer decrement using

$15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$4

together with the Cardelli et al. (1989) law (Elliott et al., 30 Jan 2026). For NGC 4666, the intrinsic Balmer ratio is taken as $15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$5 at $15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$6 (Ciraulo et al., 22 Sep 2025).

A central set of GECKOS diagnostics is built from line ratios and velocity dispersions. The survey overview defines

$15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$7

and

$15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$8

with gas-phase metallicities estimated from calibrations such as

$15\,\mathrm{Mpc}<D<70\,\mathrm{Mpc}$9

where $i>85^\circ$0 and $i>85^\circ$1 in the Pettini–Pagel prescription (Sande et al., 2023). The extraplanar-gas analyses also use

$i>85^\circ$2

BPT diagrams in $i>85^\circ$3 versus $i>85^\circ$4 with theoretical curves from Kewley et al. (2001, 2013), and shock grids from Allen et al. (2008) (Elliott et al., 30 Jan 2026).

The stellar kinematic analyses use Gauss–Hermite moments to resolve substructure. The LOSVD is parametrized as

$i>85^\circ$5

with the penalized-bias strategy explicitly tuned to recover reliable $i>85^\circ$6 in low-$i>85^\circ$7 regions (Fraser-McKelvie et al., 2024). All reduced cubes and derived two-dimensional maps are to be delivered through Data Central and ESO Phase-3 (Sande et al., 2023).

4. Stellar substructure, photometric decomposition, and dynamical modelling

GECKOS-MUSE has a strong internal focus on recovering the stellar architecture of edge-on discs through resolved kinematics. In the first 12 galaxies of the survey, 8/12 show clear boxy/peanut bulges in unsharp-masked imaging and all 8 exhibit the “double-hump” $i>85^\circ$8 profile, central $i>85^\circ$9 shoulders, and $\pm 0.3$0–$\pm 0.3$1 correlation over $\pm 0.3$2–$\pm 0.3$3, consistent with bar-buckling diagnostics (Fraser-McKelvie et al., 2024). Four galaxies—PGC 044931, NGC 3957, IC 1711, and IC 5244—show strong evidence for nuclear discs through central $\pm 0.3$4–$\pm 0.3$5 sign mismatch, “croissant”-shaped central depressions in $\pm 0.3$6 maps, strong $\pm 0.3$7 gradients, and positive $\pm 0.3$8 plateaus (Fraser-McKelvie et al., 2024). The same paper defines the nuclear-disc radius $\pm 0.3$9 at the $9.7<\log(M_\star/M_\odot)<10.3$0 turnover and reports a quantitative relation in which the strongest case, PGC 044931, has peak $9.7<\log(M_\star/M_\odot)<10.3$1 at $9.7<\log(M_\star/M_\odot)<10.3$2.

The PGC 044931 pilot study extends this structural programme by combining deep VISTA/VIRCAM $9.7<\log(M_\star/M_\odot)<10.3$3-band imaging with MUSE spectroscopy (Fraser-McKelvie et al., 19 Sep 2025). Its IMFIT decomposition yields five photometric components—an extended main disc, an edge-on stellar ring, an unresolved central point source, a boxy/peanut bulge, and an edge-on nuclear disc—plus a flat sky term, with only the extended disc, boxy/peanut bulge, and nuclear disc ever dominating more than 50% of the light in any given pixel. When these component maps are projected onto the MUSE data, the three dominant structures occupy distinct loci in the $9.7<\log(M_\star/M_\odot)<10.3$4 plane: the nuclear disc shows the classic anti-correlation $9.7<\log(M_\star/M_\odot)<10.3$5, the boxy/peanut bulge shows $9.7<\log(M_\star/M_\odot)<10.3$6, and the extended disc again shows an anti-correlation but over much larger $9.7<\log(M_\star/M_\odot)<10.3$7 (Fraser-McKelvie et al., 19 Sep 2025). In age–metallicity space, the extended disc spans $9.7<\log(M_\star/M_\odot)<10.3$8–$9.7<\log(M_\star/M_\odot)<10.3$9 and $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$00 to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$01, the boxy/peanut component is almost exclusively old and metal-poor, and the nuclear disc is $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$02–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$03 old but more metal-rich, with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$04 to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$05 (Fraser-McKelvie et al., 19 Sep 2025). The interpretation advanced in that pilot is that an extended disc experienced continued star formation, a bar formed and buckled early, and bar torques drove centrally concentrated gas inflow that built and enriched the nuclear disc.

A parallel GECKOS effort uses axisymmetric Jeans Anisotropic Models to test how well stellar dynamics can be recovered in dusty, edge-on systems (Rutherford et al., 10 Sep 2025). In the seven-galaxy JAM pilot, light and mass are described with Multi-Gaussian Expansion models and the fits are performed to

$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$06

with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$07 deemed acceptable for disc-dominated regions (Rutherford et al., 10 Sep 2025). Dust masks are constructed from stellar $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$08 maps using thresholds at $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$09, $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$10, and $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$11, labelled mild, moderate, and aggressive. In disc-dominated regions, stricter masking steadily reduces $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$12, while enclosed masses at $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$13, $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$14, and $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$15 and inclinations vary by less than 10% across all mask levels (Rutherford et al., 10 Sep 2025). Residual velocity fields reveal persistent non-axisymmetric structures in NGC 3957, IC 1711, and possibly NGC 0522, indicating that residual mapping can recover bars and boxy/peanut features even when axisymmetric JAM is used as the baseline model.

The survey overview places these analyses in a broader dynamical framework. It states that the stellar velocities and dispersions are used to build rotation curves and measure the vertical support of each component, with the simplest Jeans approximation written as

$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$16

relating the second velocity moment $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$17 and stellar density $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$18 to the gravitational potential $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$19 (Sande et al., 2023). By fitting observed $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$20 profiles at fixed radius and decomposing the surface brightness into thin and thick exponential discs, GECKOS aims to solve for scale heights, mass ratios, and intrinsic dispersions of distinct stellar components.

5. Extraplanar ionised gas, multiphase winds, and star-formation dependence

A major branch of GECKOS-MUSE is the study of extraplanar diffuse ionized gas (eDIG) and galaxy-scale winds at fixed stellar mass. The nine-galaxy star-forming subset maps extraplanar gas at $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$21–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$22 resolution and finds strong extraplanar emission reaching $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$23–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$24 from the disc midplane in all targets with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$25 (Elliott et al., 30 Jan 2026). Galaxies below $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$26 reach only $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$27–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$28, while systems with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$29 have brighter, more filamentary and more extended extraplanar H$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$30 than galaxies with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$31–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$32 (Elliott et al., 30 Jan 2026). In the 500 pc maps, high-SFR systems show bi-conical $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$33 structures with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$34 up to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$35–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$36 above the disk, whereas lower-SFR galaxies show only modest $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$37–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$38 and no clear cone (Elliott et al., 30 Jan 2026).

The same paper treats ionisation diagnostics more cautiously. In galaxy-centre regions with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$39, high-SFR systems extend from pure photoionisation into the “mixing” region and reach shock-model values with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$40–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$41 and $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$42–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$43 (Elliott et al., 30 Jan 2026). Main-sequence systems also show some spaxels entering the shock-model regime, and even higher $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$44 than the starbursts, indicating a hardened radiation field. The survey therefore reports mixed results from spatially resolved line diagnostics: the morphology of $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$45 may be useful to identify outflows, but the absolute value of the line ratio alone may not distinguish strong outflows from extraplanar gas of main-sequence galaxies (Elliott et al., 30 Jan 2026). A common misconception addressed by these results is that extraplanar optical line ratios alone cleanly separate strong winds from eDIG; GECKOS-MUSE explicitly does not make that claim.

The starburst galaxy ESO 484-036 provides the clearest resolved multiphase case within GECKOS (Hernández-Yévenes et al., 16 Apr 2026). Combining MUSE H$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$46 with ALMA CO(1–0), the study finds that both ionised and molecular gas show extraplanar emission consistent with a conical outflow; ionised gas is enclosed by molecular gas, molecular gas is detected up to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$47 from the disc, and ionised gas extends beyond $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$48 (Hernández-Yévenes et al., 16 Apr 2026). The true vertical outflow velocity is derived from

$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$49

with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$50 so that $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$51 in practice (Hernández-Yévenes et al., 16 Apr 2026). For $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$52 and $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$53–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$54, the deprojected velocities are $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$55 in both phases and are consistent with ballistic motion, with some gas possibly falling back onto the disc. The derived mass outflow rates are $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$56–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$57 and $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$58–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$59, corresponding to mass loading factors $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$60–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$61 and $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$62–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$63 (Hernández-Yévenes et al., 16 Apr 2026). Energy loading $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$64 and momentum loading $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$65 support a purely starburst-driven outflow.

NGC 4666 extends the multiphase view into the atomic regime by combining GECKOS MUSE data with WALLABY HI and ALMA/ACA CO(1–0) constraints (Ciraulo et al., 22 Sep 2025). The wind is traced in ionised and HI gas in a biconical structure extending to at least $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$66 from the galaxy disk, with increasing velocity offsets above the midplane in both phases (Ciraulo et al., 22 Sep 2025). A notable result is the resolved electron-density profile from the [S II] ratio: rather than following a simple monotonic decay, $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$67 declines from the centre to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$68, then rises again and remains high, $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$69–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$70, out to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$71 (Ciraulo et al., 22 Sep 2025). The total HI mass outflow rate above $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$72 is $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$73–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$74, while the total ionised mass outflow rate is $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$75–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$76 depending on $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$77 assumptions, and the CO outflow is constrained to $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$78 (Ciraulo et al., 22 Sep 2025). In this system, HI dominates the mass loading, and the outflow is not limited to the main bicone because a secondary starburst at the edge generates a more widespread outflow (Ciraulo et al., 22 Sep 2025).

Across these GECKOS-MUSE outflow studies, a consistent physical picture emerges: starburst galaxies host large-scale winds characterised by high $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$79, bi-conical geometry, and extended filaments, while main-sequence galaxies show eDIG more consistent with galactic fountains (Elliott et al., 30 Jan 2026). At the same time, both ESO 484-036 and NGC 4666 show velocities below the quoted escape speeds for galaxies of their mass, implying that much of the cool gas is likely to stall and rain back, thereby regulating rather than permanently quenching the gas reservoir (Hernández-Yévenes et al., 16 Apr 2026); (Ciraulo et al., 22 Sep 2025).

6. Relation to Galactic Archaeology, simulations, and survey legacy

GECKOS was conceived explicitly to extend Galactic Archaeology methods to the wider galaxy population (Sande et al., 2023). In the Milky Way, stars can be dissected into mono-abundance populations in $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$80 space and linked to age distributions and orbital properties. GECKOS generalises this paradigm by recovering local stellar age–metallicity distributions in each spaxel or Voronoi bin, identifying $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$81-rich versus $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$82-poor sequences, and constructing maps of “outer” versus “inner” disc mono-abundance populations as a function of $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$83 (Sande et al., 2023). The intended comparison space includes chemical-evolution models governed by the relative timescales of Type II and Type Ia supernova enrichment, as well as cosmological zoom-in simulations such as EAGLE, NIHAO, and FIRE (Sande et al., 2023).

The survey’s early results already sharpen several modelling problems. The NGC 4666 analysis reports that its high-$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$84 electron-density profile is at odds with classic adiabatic wind solutions or modern analytic models that predict $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$85 with $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$86–$[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$87 (Ciraulo et al., 22 Sep 2025). The ESO 484-036 study finds that including the molecular phase shifts mass-loading relations by approximately $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$88 dex relative to single-phase ionised estimates and produces a $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$89 dex discrepancy with cosmological simulations in $[\alpha/\mathrm{Fe}]-[\mathrm{Fe}/\mathrm{H}]$90, implying that current models strongly underpredict cold gas production and the role of short-range recycling flows in starburst galaxies (Hernández-Yévenes et al., 16 Apr 2026). The nine-galaxy extraplanar-gas paper also invokes TNG50 predictions of persistent high-ionisation cones from past AGN outbursts as a possible explanation for relic features in low-SFR galaxies such as IC 1711 (Elliott et al., 30 Jan 2026).

On the stellar-dynamical side, GECKOS-MUSE places limits on what simplified models can and cannot recover. The JAM pilot demonstrates that global dynamical parameters such as enclosed mass and inclination are stable to within 10% across dust masks, but it also emphasises that the axisymmetry assumption breaks down in central bar/bulge regions, that mismatches between IR light and optical kinematics can mis-model dynamically cold components, and that future work will require radiative-transfer dust modelling and orbit-superposition Schwarzschild models for explicit bar dynamics (Rutherford et al., 10 Sep 2025). A plausible implication is that GECKOS functions not just as a survey of nearby edge-on galaxies, but also as a methodological testbed for disentangling dust, projection effects, and non-axisymmetric structure in integral-field spectroscopy.

The longer-term legacy of the survey lies in its combined chemodynamical and multiphase perspective. GECKOS is designed to deliver spatially resolved measurements of stellar abundances, ages, and kinematics together with ionised-gas metallicities, ionisation parameters, pressure, and inflow and outflow kinematics (Sande et al., 2023). Later papers show that this framework can be specialised toward kinematic substructure, structural decomposition, eDIG, and resolved atomic, molecular, and ionised outflows without abandoning the mass-controlled logic of the parent sample (Fraser-McKelvie et al., 2024); (Elliott et al., 30 Jan 2026). Taken together, the published GECKOS-MUSE results suggest a continuous transition from fountain-dominated extraplanar layers in main-sequence systems to shock- and wind-driven bi-cones in starbursts, while also indicating that bar-driven secular evolution, nuclear-disc growth, and vertical thickening are pervasive in the stellar body of edge-on discs (Elliott et al., 30 Jan 2026).